REDUCING THE GENDER GAP IN FIRST-YEAR PHYSICS PERFORMANCE

Abstract

Multiple choice tests have been the main means of assessing Newtonian mechanics in the first year physics course at UNSW Canberra for several years. A study of persistent gender gaps on these tests led us to perform a question-by-question analysis of test results for the years 2010-2015. In 2016, the course changed its weekly contact pattern from three hours of lectures supplemented by a one-hour tutorial, to two one-hour lectures and a two-hour workshop-tutorial. The mechanics test was also changed to consist of a reduced number of multiple choice questions (MCQs), and to include four written answer problems. Of these four problems, three had previously been used as MCQs. All three showed smaller gender gaps than the MCQ versions of the same questions. Two MCQs previously identified as showing large, persistent gaps were modified in an attempt to reduce the gap. One of these showed a significantly reduced gap and the other an unchanged gap. Ten previously used MCQs were repeated, and showed a small average decrease in gap. We ascribe this overall reduction in the gender gap to better communication: the more interactive teaching methods improved student-student and student-teacher interactions; and the re-casting of assessment questions gave students more means of communicating their understanding to us. However we note that this improvement comes at the ongoing cost of increased staff time for both facilitating tutorial work and marking.